Osteoporosis is associated with bone loss, both cortical and trabecular, eventually resulting in structural failure. Most fractures occur at sites of a high degree of trabecular bone (vertebral proximal femur, distal radius, etc.). For this reason, the focus of osteoporosis has been on density of trabecular bone, as measured by bone mineral densitometry. The unsatisfactory outcome prediction of bone density prompted the investigators to search for alternative approaches relying on an assessment of structure rather than material density. The hypotheses underlying this project proposal are based on the spatially nonuniform magnetic field induced by the different magnetic permeability of bone and bone marrow, with the amplitude and direction of the induced field being a function of trabecular morphology and thus trabecular bone strength. The relative magnetic field inhomogeneity has been shown to be quantifiable in terms ofR2', the contribution from the induced inhomogeneous field to the total rate of decay of the transverse magnetization R. The principal aim of this proposal is to apply substantially improved methods to both the proximal femur and lumbar spine, in vitro in human cadaver specimens and in vivo in osteoporosis patients and their controls. The hip is a major osteoporotic fracture site. It is biomechanically more complex because of the multidirectional nature of the stresses to which it is subjected. Besides incorporation of new measurement methods providing higher precision, algorithms will be implemented for an exact measurement of the fat volume fraction as a means to correct the R2* data for the different magnetic susceptibility of fat and water relative to bone. It is expected that these steps will lead to improved diagnostic accuracy. As a gold standard for the presence or absence of osteoporosis objective criteria of vertebral deformity will be applied. By extending the technique to the hip, it is intended to demonstrate the more general validity of the investigators' previous results which proved a relationship between the induced magnetic field and trabecular structural parameters including number density, orientation and thickness; and the bone's modulus of elasticity and thus strength of the lumbar vertebrae. Specifically, it is proposed to assess in vitro whether Young's modulus for axial loading scales with R2', and whether in vivo the changes in bone remodeling occurring during aging and osteoporosis parallel those in the spine. Finally, the diagnostic accuracy of the method will be compared to conventional bone densitometry.